zilliztech / zilliztech/VectorDBBench

VectorDBBench running on Milvus DB giving error - Resource temporarily unavailable

Open
#368 8 comments 0 reactions 1 assignee View on GitHub

@alwayslove2013 is already working on this.

Since Sep 18, 2024.

Dominant language
Python
Stars
1.2k
Forks
438
Avg merge
1d 4h
Merged PRs (30d)
1

Description

We are running a VectorDB benchmark test on Milvus DB. Below are the specification of the test ran-
Database - Milvus
Case - Search Performance Test (500K Dataset, 1536 Dim)
Index type - DiskANN
search_list - 100 (default)

Sharing the full log here-

#init_bench
2024-09-17 09:59:38,437 | INFO: all configs: [('ALIYUN_OSS_URL', 'assets.zilliz.com.cn/benchmark/'), ('AWS_S3_URL', 'assets.zilliz.com/benchmark/'), ('CONCURRENCY_DURATION', 30), ('CONFIG_LOCAL_DIR', PosixPath('/usr/local/lib/python3.11/site-packages/vectordb_bench/config-files')), ('CUSTOM_CONFIG_DIR', PosixPath('/usr/local/lib/python3.11/site-packages/vectordb_bench/custom/custom_case.json')), ('DATASET_LOCAL_DIR', PosixPath('/tmp/vectordb_bench/dataset')), ('DEFAULT_DATASET_URL', 'assets.zilliz.com/benchmark/'), ('DROP_OLD', True), ('K_DEFAULT', 100), ('LOG_LEVEL', 'INFO'), ('NUM_CONCURRENCY', [1, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100]), ('NUM_PER_BATCH', 5000), ('RESULTS_LOCAL_DIR', PosixPath('/usr/local/lib/python3.11/site-packages/vectordb_bench/results')), ('USE_SHUFFLED_DATA', True)] (main.py:11) (15519)

Collecting usage statistics. To deactivate, set browser.gatherUsageStats to false.

You can now view your Streamlit app in your browser.

Local URL: http://localhost:8501
Network URL: http://10.129.4.79:8501
External URL: http://192.133.242.70:8501

2024-09-17 10:01:34,491 | INFO: generated uuid for the tasks: e2aaa4749e1b41499c63d75622b32a74 (interface.py:66) (15520)
2024-09-17 10:01:34,552 | INFO | DB | CaseType Dataset Filter | task_label (task_runner.py:338)
2024-09-17 10:01:34,552 | INFO | ----------- | ------------ -------------------- ------- | ------- (task_runner.py:338)
2024-09-17 10:01:34,552 | INFO | Milvus | Performance OpenAI-MEDIUM-500K None | test2_nfs_17 (task_runner.py:338)
2024-09-17 10:01:34,552 | INFO: task submitted: id=e2aaa4749e1b41499c63d75622b32a74, test2_nfs_17, case number: 1 (interface.py:231) (15520)
2024-09-17 10:01:34.775 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:01:35,220 | INFO: [1/1] start case: {'label': <CaseLabel.Performance: 2>, 'dataset': {'data': {'name': 'OpenAI', 'size': 500000, 'dim': 1536, 'metric_type': <MetricType.COSINE: 'COSINE'>}}, 'db': 'Milvus'}, drop_old=True (interface.py:164) (15635)
2024-09-17 10:01:35,220 | INFO: Starting run (task_runner.py:100) (15635)
2024-09-17 10:01:35,297 | INFO: Milvus client drop_old collection: VectorDBBenchCollection (milvus.py:45) (15635)
2024-09-17 10:01:35,302 | INFO: Milvus create collection: VectorDBBenchCollection (milvus.py:55) (15635)
2024-09-17 10:01:35,950 | INFO: local dataset root path not exist, creating it: /tmp/vectordb_bench/dataset/openai/openai_medium_500k (data_source.py:119) (15635)
2024-09-17 10:01:35,950 | INFO: Start to downloading files, total count: 3 (data_source.py:135) (15635)
100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [02:20<00:00, 46.86s/it]
2024-09-17 10:03:56,531 | INFO: Succeed to download all files, downloaded file count = 3 (data_source.py:140) (15635)
2024-09-17 10:03:56,531 | INFO: Read the entire file into memory: test.parquet (dataset.py:229) (15635)
2024-09-17 10:03:56,665 | INFO: Read the entire file into memory: neighbors.parquet (dataset.py:229) (15635)
2024-09-17 10:03:56,690 | INFO: Start performance case (task_runner.py:158) (15635)
2024-09-17 10:03:56.912 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:03:57,480 | INFO: (SpawnProcess-1:1) Start inserting embeddings in batch 5000 (serial_runner.py:35) (15962)
2024-09-17 10:03:57,480 | INFO: Get iterator for shuffle_train.parquet (dataset.py:247) (15962)
2024-09-17 10:04:37,844 | INFO: (SpawnProcess-1:1) Loaded 100000 embeddings into VectorDB (serial_runner.py:59) (15962)
2024-09-17 10:05:20,985 | INFO: (SpawnProcess-1:1) Loaded 200000 embeddings into VectorDB (serial_runner.py:59) (15962)
2024-09-17 10:06:12,291 | INFO: (SpawnProcess-1:1) Loaded 300000 embeddings into VectorDB (serial_runner.py:59) (15962)
2024-09-17 10:07:06,173 | INFO: (SpawnProcess-1:1) Loaded 400000 embeddings into VectorDB (serial_runner.py:59) (15962)
2024-09-17 10:07:58,594 | INFO: (SpawnProcess-1:1) Loaded 500000 embeddings into VectorDB (serial_runner.py:59) (15962)
2024-09-17 10:07:58,603 | INFO: (SpawnProcess-1:1) Finish loading all dataset into VectorDB, dur=241.12337821099936 (serial_runner.py:61) (15962)
2024-09-17 10:07:59.174 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:08:00,333 | INFO: Milvus post insert before optimize (milvus.py:101) (18207)
2024-09-17 10:18:08,647 | INFO: Milvus optimizing before search (milvus.py:93) (18207)
2024-09-17 10:18:18,257 | INFO: Finish loading the entire dataset into VectorDB, insert_duration=242.15658646000156, optimize_duration=617.8485961160004 load_duration(insert + optimize) = 860.0052 (task_runner.py:167) (15635)
2024-09-17 10:18:18.525 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:18:19,117 | INFO: SpawnProcess-1:3 start search the entire test_data to get recall and latency (serial_runner.py:171) (19339)
2024-09-17 10:18:52,124 | INFO: SpawnProcess-1:3 search entire test_data: cost=25.8579s, queries=1000, avg_recall=0.9896, avg_ndcg=0.9911,avg_latency=0.0259, p99=0.0557 (serial_runner.py:209) (19339)
2024-09-17 10:18:52.526 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:18:52,972 | INFO: Start search 30s in concurrency 1, filters: None (mp_runner.py:99) (15635)
2024-09-17 10:18:53.196 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing
2024-09-17 10:19:29,975 | INFO: Syncing all process and start concurrency search, concurrency=5 (mp_runner.py:108) (15635)
2024-09-17 10:19:59,998 | INFO: SpawnProcess-1:8 search 30s: actual_dur=30.0072s, count=1636, qps in this process: 54.5202 (mp_runner.py:76) (20009)
2024-09-17 10:20:00,006 | INFO: SpawnProcess-1:7 search 30s: actual_dur=30.0141s, count=1626, qps in this process: 54.1745 (mp_runner.py:76) (20006)
2024-09-17 10:20:00,008 | INFO: SpawnProcess-1:11 search 30s: actual_dur=30.0159s, count=1629, qps in this process: 54.2712 (mp_runner.py:76) (20012)
2024-09-17 10:20:00,010 | INFO: SpawnProcess-1:10 search 30s: actual_dur=30.0189s, count=1631, qps in this process: 54.3324 (mp_runner.py:76) (20011)
2024-09-17 10:20:00,011 | INFO: SpawnProcess-1:9 search 30s: actual_dur=30.0186s, count=1630, qps in this process: 54.2997 (mp_runner.py:76) (20010)
2024-09-17 10:20:00,013 | INFO: End search in concurrency 5: dur=30.037750658000732s, total_count=8152, qps=271.3918 (mp_runner.py:120) (15635)
2024-09-17 10:20:00,362 | INFO: Update largest qps with concurrency 5: current max_qps=271.3918 (mp_runner.py:124) (15635)
2024-09-17 10:20:00.585 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:20:01,031 | INFO: Start search 30s in concurrency 10, filters: None (mp_runner.py:99) (15635)
2024-09-17 10:20:11,068 | INFO: Syncing all process and start concurrency search, concurrency=10 (mp_runner.py:108) (15635)
2024-09-17 10:20:41,098 | INFO: SpawnProcess-1:18 search 30s: actual_dur=30.011s, count=885, qps in this process: 29.4892 (mp_runner.py:76) (20990)
2024-09-17 10:20:41,099 | INFO: SpawnProcess-1:13 search 30s: actual_dur=30.0114s, count=887, qps in this process: 29.5554 (mp_runner.py:76) (20983)
2024-09-17 10:20:41,104 | INFO: SpawnProcess-1:15 search 30s: actual_dur=30.0177s, count=895, qps in this process: 29.8157 (mp_runner.py:76) (20987)
2024-09-17 10:20:41,104 | INFO: SpawnProcess-1:17 search 30s: actual_dur=30.0173s, count=892, qps in this process: 29.7162 (mp_runner.py:76) (20989)
2024-09-17 10:20:41,104 | INFO: SpawnProcess-1:21 search 30s: actual_dur=30.0171s, count=891, qps in this process: 29.6831 (mp_runner.py:76) (20993)
2024-09-17 10:20:41,105 | INFO: SpawnProcess-1:22 search 30s: actual_dur=30.0176s, count=892, qps in this process: 29.7159 (mp_runner.py:76) (20994)
2024-09-17 10:20:41,106 | INFO: SpawnProcess-1:20 search 30s: actual_dur=30.0184s, count=890, qps in this process: 29.6485 (mp_runner.py:76) (20992)
2024-09-17 10:20:41,106 | INFO: SpawnProcess-1:19 search 30s: actual_dur=30.0193s, count=894, qps in this process: 29.7808 (mp_runner.py:76) (20991)
2024-09-17 10:20:41,106 | INFO: SpawnProcess-1:16 search 30s: actual_dur=30.0177s, count=885, qps in this process: 29.4826 (mp_runner.py:76) (20988)
2024-09-17 10:20:41,106 | INFO: SpawnProcess-1:14 search 30s: actual_dur=30.0178s, count=890, qps in this process: 29.6491 (mp_runner.py:76) (20986)
2024-09-17 10:20:41,108 | INFO: End search in concurrency 10: dur=30.040207265999925s, total_count=8901, qps=296.3029 (mp_runner.py:120) (15635)
2024-09-17 10:20:41,512 | INFO: Update largest qps with concurrency 10: current max_qps=296.3029 (mp_runner.py:124) (15635)
2024-09-17 10:20:41.734 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:20:42,178 | INFO: Start search 30s in concurrency 15, filters: None (mp_runner.py:99) (15635)
2024-09-17 10:20:42.420 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:20:52,215 | INFO: Syncing all process and start concurrency search, concurrency=15 (mp_runner.py:108) (15635)
2024-09-17 10:21:22,247 | INFO: SpawnProcess-1:31 search 30s: actual_dur=30.0064s, count=578, qps in this process: 19.2626 (mp_runner.py:76) (22862)
2024-09-17 10:21:22,248 | INFO: SpawnProcess-1:27 search 30s: actual_dur=30.005s, count=578, qps in this process: 19.2635 (mp_runner.py:76) (22858)
2024-09-17 10:21:22,249 | INFO: SpawnProcess-1:38 search 30s: actual_dur=30.009s, count=577, qps in this process: 19.2276 (mp_runner.py:76) (22869)
2024-09-17 10:21:22,250 | INFO: SpawnProcess-1:34 search 30s: actual_dur=30.0074s, count=581, qps in this process: 19.3619 (mp_runner.py:76) (22865)
2024-09-17 10:21:22,253 | INFO: SpawnProcess-1:36 search 30s: actual_dur=30.0156s, count=574, qps in this process: 19.1234 (mp_runner.py:76) (22867)
2024-09-17 10:21:22,254 | INFO: SpawnProcess-1:24 search 30s: actual_dur=30.0121s, count=576, qps in this process: 19.1923 (mp_runner.py:76) (22853)
2024-09-17 10:21:22,255 | INFO: SpawnProcess-1:29 search 30s: actual_dur=30.012s, count=577, qps in this process: 19.2256 (mp_runner.py:76) (22860)
2024-09-17 10:21:22,256 | INFO: SpawnProcess-1:33 search 30s: actual_dur=30.0164s, count=581, qps in this process: 19.3561 (mp_runner.py:76) (22864)
2024-09-17 10:21:22,260 | INFO: SpawnProcess-1:32 search 30s: actual_dur=30.0173s, count=584, qps in this process: 19.4554 (mp_runner.py:76) (22863)
2024-09-17 10:21:22,261 | INFO: SpawnProcess-1:28 search 30s: actual_dur=30.0191s, count=587, qps in this process: 19.5542 (mp_runner.py:76) (22859)
2024-09-17 10:21:22,263 | INFO: SpawnProcess-1:25 search 30s: actual_dur=30.0217s, count=583, qps in this process: 19.4193 (mp_runner.py:76) (22856)
2024-09-17 10:21:22,264 | INFO: SpawnProcess-1:26 search 30s: actual_dur=30.0245s, count=576, qps in this process: 19.1843 (mp_runner.py:76) (22857)
2024-09-17 10:21:22,264 | INFO: SpawnProcess-1:37 search 30s: actual_dur=30.0236s, count=587, qps in this process: 19.5513 (mp_runner.py:76) (22868)
2024-09-17 10:21:22,264 | INFO: SpawnProcess-1:35 search 30s: actual_dur=30.0258s, count=581, qps in this process: 19.35 (mp_runner.py:76) (22866)
2024-09-17 10:21:22,266 | INFO: SpawnProcess-1:30 search 30s: actual_dur=30.0242s, count=576, qps in this process: 19.1845 (mp_runner.py:76) (22861)
2024-09-17 10:21:22,267 | INFO: End search in concurrency 15: dur=30.05192263400022s, total_count=8696, qps=289.3658 (mp_runner.py:120) (15635)
2024-09-17 10:21:22.933 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:21:23,382 | INFO: Start search 30s in concurrency 20, filters: None (mp_runner.py:99) (15635)
2024-09-17 10:21:23.624 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:21:23.705 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing
2024-09-17 10:21:33,434 | INFO: Syncing all process and start concurrency search, concurrency=20 (mp_runner.py:108) (15635)
2024-09-17 10:22:03,471 | INFO: SpawnProcess-1:40 search 30s: actual_dur=30.0123s, count=437, qps in this process: 14.5607 (mp_runner.py:76) (25609)
2024-09-17 10:22:03,472 | INFO: SpawnProcess-1:50 search 30s: actual_dur=30.0141s, count=435, qps in this process: 14.4932 (mp_runner.py:76) (25621)
2024-09-17 10:22:03,474 | INFO: SpawnProcess-1:54 search 30s: actual_dur=30.0157s, count=433, qps in this process: 14.4258 (mp_runner.py:76) (25625)
2024-09-17 10:22:03,474 | INFO: SpawnProcess-1:59 search 30s: actual_dur=30.0168s, count=435, qps in this process: 14.4919 (mp_runner.py:76) (25630)
2024-09-17 10:22:03,475 | INFO: SpawnProcess-1:44 search 30s: actual_dur=30.0151s, count=438, qps in this process: 14.5927 (mp_runner.py:76) (25615)
2024-09-17 10:22:03,476 | INFO: SpawnProcess-1:45 search 30s: actual_dur=30.018s, count=435, qps in this process: 14.4913 (mp_runner.py:76) (25616)
2024-09-17 10:22:03,478 | INFO: SpawnProcess-1:58 search 30s: actual_dur=30.0182s, count=433, qps in this process: 14.4246 (mp_runner.py:76) (25629)
2024-09-17 10:22:03,480 | INFO: SpawnProcess-1:49 search 30s: actual_dur=30.0205s, count=436, qps in this process: 14.5234 (mp_runner.py:76) (25620)
2024-09-17 10:22:03,482 | INFO: SpawnProcess-1:51 search 30s: actual_dur=30.0238s, count=440, qps in this process: 14.655 (mp_runner.py:76) (25622)
2024-09-17 10:22:03,483 | INFO: SpawnProcess-1:57 search 30s: actual_dur=30.0271s, count=439, qps in this process: 14.6201 (mp_runner.py:76) (25628)
2024-09-17 10:22:03,486 | INFO: SpawnProcess-1:48 search 30s: actual_dur=30.0287s, count=439, qps in this process: 14.6193 (mp_runner.py:76) (25619)
2024-09-17 10:22:03,488 | INFO: SpawnProcess-1:46 search 30s: actual_dur=30.0304s, count=435, qps in this process: 14.4853 (mp_runner.py:76) (25617)
2024-09-17 10:22:03,489 | INFO: SpawnProcess-1:53 search 30s: actual_dur=30.0343s, count=435, qps in this process: 14.4834 (mp_runner.py:76) (25624)
2024-09-17 10:22:03,493 | INFO: SpawnProcess-1:43 search 30s: actual_dur=30.0324s, count=440, qps in this process: 14.6508 (mp_runner.py:76) (25614)
2024-09-17 10:22:03,494 | INFO: SpawnProcess-1:52 search 30s: actual_dur=30.0388s, count=438, qps in this process: 14.5811 (mp_runner.py:76) (25623)
2024-09-17 10:22:03,498 | INFO: SpawnProcess-1:47 search 30s: actual_dur=30.0409s, count=434, qps in this process: 14.447 (mp_runner.py:76) (25618)
2024-09-17 10:22:03,498 | INFO: SpawnProcess-1:55 search 30s: actual_dur=30.0375s, count=433, qps in this process: 14.4153 (mp_runner.py:76) (25626)
2024-09-17 10:22:03,498 | INFO: SpawnProcess-1:41 search 30s: actual_dur=30.0386s, count=439, qps in this process: 14.6145 (mp_runner.py:76) (25612)
2024-09-17 10:22:03,499 | INFO: SpawnProcess-1:42 search 30s: actual_dur=30.0421s, count=432, qps in this process: 14.3798 (mp_runner.py:76) (25613)
2024-09-17 10:22:03,499 | INFO: SpawnProcess-1:56 search 30s: actual_dur=30.0395s, count=437, qps in this process: 14.5475 (mp_runner.py:76) (25627)
2024-09-17 10:22:03,501 | INFO: End search in concurrency 20: dur=30.067121991000022s, total_count=8723, qps=290.1176 (mp_runner.py:120) (15635)
2024-09-17 10:22:04.157 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:22:04,613 | INFO: Start search 30s in concurrency 25, filters: None (mp_runner.py:99) (15635)
2024-09-17 10:22:04.868 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:22:04.868 WARNING streamlit.runtime.scriptrunner_utils.script_run_context: Thread 'MainThread': missing ScriptRunContext! This warning can be ignored when running in bare mode.
2024-09-17 10:22:14,655 | INFO: Syncing all process and start concurrency search, concurrency=25 (mp_runner.py:108) (15635)
OpenBLAS blas_thread_init: pthread_create failed for thread 22 of 64: Resource temporarily unavailable
OpenBLAS blas_thread_init: RLIMIT_NPROC 4194304 current, 4194304 max
OpenBLAS blas_thread_init: pthread_create failed for thread 23 of 64: Resource temporarily unavailable
OpenBLAS blas_thread_init: RLIMIT_NPROC 4194304 current, 4194304 max

Usingpymilvus, we validated that the db connection, creation of collections and indexing are successful.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.